To address the need for non-pharmacological treatment in young children with ADHD through a test of specific and non-specific efficacy of a promising UK home-based intervention [New Forest Parenting Package (NFPP)]. Rationale: ADHD originates in early childhood and predicts multiple functional impairments. Although psychostimulants are effective in preschoolers with ADHD, families resist their use. Therefore, establishing effective psychosocial interventions for preschoolers represents a public health priority. NFPP is the only intervention with documented immediate and sustained efficacy for ADHD and ODD symptoms in preschoolers. The study evaluates NFPP's superiority to a standard parenting intervention [Helping Non-Compliant Children (HNCC)] and a wait-list (WL) control. HNCC has been shown to improve oppositional behaviors in preschoolers. Its effect on ADHD symptoms is untested. Hypotheses: In 3 year olds with ADHD, NFPP is superior to HNCC and WL in reducing ADHD symptoms, and children's delay tolerance for rewards;NFPP is superior to WL in reducing ODD symptoms;NFPP's superiority over HNCC on ADHD symptoms will be maintained into the school year following the intervention. We will explore the moderating effect of caretaker's ADHD on treatment outcome, and whether changes in parenting mediate treatment efficacy in NFPP and HNCC. Design: 225 preschoolers (3.0 - 3-11) with ADHD at home and school will be randomized to 1) NFPP, 2) HNCC, or 3) WL for 8 weeks. ADHD measures include ratings by primary caretaker, teacher, clinician and direct child observations. Caretakers will be assessed for ADHD diagnosis and general psychopathology. Multi-informant, multi-methods, including observations of parent- child interactions, will assess parental behavior and children's function at the end of treatment, and into the fall of the school year following treatment. Analyses: ITT ANCOVA's will test immediate and follow-up treatment differences. Statistical interactions will explore relationships between postulated moderators (i.e., caretaker ADHD) and treatment differences. Structural equation modeling will explore the mediating role of changes in caretaker behaviors.